{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/text2mol-cross-modal-molecule-retrieval-with","title":"Text2Mol: Cross-Modal Molecule Retrieval with Natural Language Queries","arxiv_id":null,"date":"2021-11-01","proceeding":"EMNLP 2021 11","authors":["Carl Edwards","ChengXiang Zhai","Heng Ji"],"abstract":"We propose a new task, Text2Mol, to retrieve molecules using natural language descriptions as queries. Natural language and molecules encode information in very different ways, which leads to the exciting but challenging problem of integrating these two very different modalities. Although some work has been done on text-based retrieval and structure-based retrieval, this new task requires integrating molecules and natural language more directly. Moreover, this can be viewed as an especially challenging cross-lingual retrieval problem by considering the molecules as a language with a very unique grammar. We construct a paired dataset of molecules and their corresponding text descriptions, which we use to learn an aligned common semantic embedding space for retrieval. We extend this to create a cross-modal attention-based model for explainability and reranking by interpreting the attentions as association rules. We also employ an ensemble approach to integrate our different architectures, which significantly improves results from 0.372 to 0.499 MRR. This new multimodal approach opens a new perspective on solving problems in chemistry literature understanding and molecular machine learning.","url_abs":"https://aclanthology.org/2021.emnlp-main.47","url_pdf":"https://aclanthology.org/2021.emnlp-main.47.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"text2mol-cross-modal-molecule-retrieval-with","repo_url":"https://github.com/cnedwards/text2mol","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"cross-modal-retrieval","task_name":"Cross-Modal Retrieval"},{"task_slug":"natural-language-queries","task_name":"Natural Language Queries"},{"task_slug":"reranking","task_name":"Reranking"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[{"slug":"chebi-20","name":"ChEBI-20","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/cross-modal-retrieval-on-chebi-20","task":"Cross-Modal Retrieval","dataset":"ChEBI-20","model":"All-Ensemble","rank_in_archive_order":7,"of":9,"metrics":{"Hits@1":"34.4","Hits@10":"81.1","Mean Rank":"20.21","Test MRR":"49.9"},"uses_additional_data":false},{"leaderboard":"/sota/cross-modal-retrieval-on-chebi-20","task":"Cross-Modal Retrieval","dataset":"ChEBI-20","model":"MLP1","rank_in_archive_order":8,"of":9,"metrics":{"Hits@1":"22.4","Hits@10":"68.6","Mean Rank":"30.38","Test MRR":"37.2"},"uses_additional_data":false},{"leaderboard":"/sota/cross-modal-retrieval-on-chebi-20","task":"Cross-Modal Retrieval","dataset":"ChEBI-20","model":"GCN2","rank_in_archive_order":9,"of":9,"metrics":{"Hits@1":"22.3","Hits@10":"68.9","Mean Rank":"41.90","Test MRR":"37.1"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}